The Portal Tax Just Got Wider

Last Tuesday’s federal Budget already had property sitting right in the middle of the national argument, with proposed changes to negative gearing and capital gains tax giving agents, investors, vendors and homeowners plenty to think about. Then on Wednesday morning, realestate.com.au sent through its FY27 residential pricing notice to agents, written in the usual careful language of “updated pricing”, “price changes”, “flexibility” and “performance”. Strip away the softer wording, though, and the direction appears familiar enough: the cost of advertising property on realestate.com.au is moving again.

This is more than an annual supplier increase. The pricing notice sits across a much bigger argument about portal pricing power, vendor-paid advertising, opaque suburb-based pricing, diluted depth products, and the growing number of ways agencies may eventually start working around the portal model. That has been part of my thinking for a while, but the timing of this email made the issue feel much more immediate.

As many of you know, I have been shifting more of my attention toward Real Estate AIM and away from the day-to-day agency side of frontline sales. More on that later. So part of me could simply shrug and say this particular increase will not sit on my desk for long, but that is also why I would rather say it plainly now. This is not a normal supplier putting up prices in a tight market. This is what pricing power looks like when a business knows the industry has been trained to treat its product as close to unavoidable.

The expected Premiere+ movement was annoying enough, mainly because that has become grimly predictable. For the suburbs I mainly work in, REA tells me pricing will move from $1,379 to $1,519 from 1 July 2026, an increase of $140, or just over 10%. Other suburbs with significantly smaller populations in my region, move from $1,219 to $1,319, an increase of $100, or a little over 8%. That is local pricing rather than a national average, but it is useful because it shows what the increase looks like at ground level.

Those numbers sit on top of the other confirmed change, which is the move from $0 to $169 for regional Standard Land and Rural listings that previously sat at no charge under my current Premiere+ All Buy contract. You cannot really describe a move from zero as a normal percentage increase, because the starting point was nothing. Something that was included free of charge now has a price attached to it.

For a metro office, land and rural may sound like the edges of the business. In regional markets, they are not. Plenty of offices carry a meaningful split across residential, land and rural stock. If those listings now attract a Standard listing charge where they previously sat at $0 under the relevant account structure, this becomes a material new annual cost, not a rounding error or an admin tweak.

It also changes a practical sales conversation. Previously, if a rural or land listing did not justify a paid depth product, the agent could still get it online as a Standard listing at no additional listing cost under that contract structure. That gave the agent a bit of flexibility with a vendor who did not want to spend heavily, or where the property type, likely buyer pool, or campaign strategy did not justify the higher paid product. From FY27, even that fallback is no longer free.

The choice then becomes awkward in a very ordinary, practical way. An agency can absorb the $169 as a small but repeated cost of doing business, or start charging vendors for something that was previously included and therefore never needed to be sold. On one listing, $169 may not sound dramatic. Across a regional office that handles land and rural stock regularly, it becomes another line of leakage. It is also exactly the sort of charge that is irritating enough to matter, but fiddly enough that many agents may quietly absorb it rather than have yet another advertising conversation with an owner.

I keep coming back to the word tax because, commercially, that is how it behaves. It is not a government tax, obviously, but for many sellers a realestate.com.au listing does not feel optional, and for many agents, especially those on an All style depth contract, there is no simple ability to say they will just skip that product this time. Either the vendor pays or the agency absorbs it, while the portal still gets paid. REA does not have to sit in the lounge room explaining the increase to the owner. The agent does.

A supplier only gets that kind of leverage when the product has become close to unavoidable. REA has tens of thousands of agents effectively selling its product on its behalf, explaining its pricing, defending its value and dealing with the owner’s frustration when the cost becomes difficult to justify. In my opinion, if REA had only 10% of the market, it could not behave this way. If the market had three genuinely even players, each with a third of buyer attention and vendor confidence, pricing would be on the menu in a very different way, because a competitor would use price to take share, a challenger would force restraint, and a supplier would think twice before widening the base of what it charges for. But that is not the market agents are operating in, and it is certainly not the market most vendors experience when the advertising schedule is put in front of them.

REA will say the audience is real, and that is fair enough. The platform does have value, and nobody serious is pretending otherwise. Its Buyer Impact Model says realestate.com.au attracts and engages the buyer on 9 in 10 properties listed on the platform that go on to sell, which is clearly a serious claim and one REA has put work into supporting. But the existence of value does not mean the market should treat every increase as justified.

REA’s public numbers make the issue sharper, although the wording needs unpacking because “yield” can sound more technical than it really is. In REA’s Q3 FY26 update, Australian residential revenue was up 12%. Its Buy revenue growth was driven by a 14% increase in Buy yield and only a 1% increase in national listings. In plain English, Buy yield is the amount of revenue REA is extracting from the residential sales-listing side of the business. It is not buyer activity, it is not simply audience growth, it is effectively the monetisation of the Buy listing base.

REA says that 14% Buy yield increase was helped by a 7% average Premiere+ price rise, growth in add-on products, higher subscription revenues, increased depth penetration and a small positive impact from geographical mix. So the 7% figure is not the full story, it is the average Premiere+ price rise REA disclosed, sitting inside a broader 14% uplift in how much money the Buy business extracted per listing.

If listing volumes were up only 1%, but Buy yield was up 14%, the growth is not mainly coming from more properties being advertised. It is coming from REA charging more, selling more depth, adding more products, increasing subscription revenue and pushing more of the market into higher-priced settings. The machine is not just getting bigger. It is getting better at extracting more from each listing that passes through it.

The “average” Premiere+ price rise can also be misleading at agency level. A 7% average across the network does not mean every agency, every suburb or every regional market is seeing 7%. In the local examples REA has now provided to me, the increase is roughly 8.2% for locations like Sandy Point and Waratah Bay, and roughly 10% for townships Foster and Mardan. Again, that is not a national claim, it is a local one, but it does show why the lived experience of agents can be very different from the softer average number.

REA’s own result makes this look less like a business recovering higher costs and more like a business lifting yield from a market it already dominates. Nobody serious thinks a platform of this scale runs for free. Server costs, product development, data, support, AI, sales, compliance and marketing all exist. But the marginal cost of serving another listing has not doubled in five years, and the cost of putting a land or rural listing online has not suddenly moved so dramatically that another charge line becomes unavoidable. If anything, many businesses of this scale are constantly looking for operational efficiencies, including centralised service, lower human support load and more automated customer handling. This does not look like a cost-pressure story, it looks like a pricing-power story.

The pricing structure itself probably deserves more attention than it gets. REA’s public help material says the cost of advertising a listing varies according to the agency’s subscription, the upgrade purchased and the location of the property. It also says residential subscribers access their contracted Premiere pricing by entering their Agency ID into the Suburb Pricing Guide. Standard listing pricing and other upgrade pricing also sit behind Ignite or suburb pricing tools, which means there is no simple national public rate card that a vendor can look at and understand across the country.

That hidden, localised pricing becomes important when you start thinking about who is actually paying the bill. A vendor in Toorak may be paying several times what a vendor in Leongatha pays for a broadly similar listing product, even where the property value may be comparable. If a $2 million house in one suburb costs materially more to list than a $2 million acreage property in another, what exactly is being priced? The cost of delivery has not changed in any meaningful way. The screen space is not more expensive. The listing does not require some radically different infrastructure. The difference appears to be location, contract and what REA believes the market will bear.

I am not making a legal finding, and the ACCC can speak for itself. Commercially, though, a pricing structure like this should be much easier for vendors to understand. If the end payer cannot easily tell whether they are paying for value, scarcity, market power, or postcode-based extraction, the structure itself becomes part of the problem.

The suburb-based pricing issue becomes even stranger once the ranking mechanics are taken into account. A vendor in a higher-priced metro suburb may be asked to pay far more for the same broad advertising tier than a vendor in a lower-priced regional suburb, but the underlying question is not just whether the metro property sits in a higher-value market, the better question is whether the extra cost buys a proportionate increase in actual advantage.

REA would no doubt argue that the city listing sits in a higher-value market, attracts more buyer attention, and therefore carries a higher advertising value. That is the predictable commercial defence, but the more you look at the mechanics of the product, the more strained that argument becomes. The underlying platform is the same, the listing infrastructure is the same, the search logic is the same. The alerts, branding, campaign tools and consumer-facing experience are broadly the same. The cost to serve one listing in Toorak is not three times the cost to serve one listing in Leongatha in any meaningful operational sense. What changes is the market REA is selling into, and therefore what it believes it can charge.

There is another oddity in the value case. In a low-volume regional suburb with only say two or three (or even 10) active listings, the incremental value of paying for a premium position can be very thin, because if a serious buyer is looking in that town, they are likely to see all of the available properties anyway. Being “at the top” of a handful of listings is not worth anything like the same amount as being prominent in a crowded metro market. So if an agency is contractually required to load that property as Premiere+, the vendor may be paying for a premium position they barely need.

In a high-volume suburb, the opposite problem appears. The vendor may pay much more, but they are also competing against many more properties inside the same paid tier. If 20, 30 or 40 listings in Ringwood, Camberwell, Hawthorn, Toorak, Berwick or Richmond are all paying for Premiere or Premiere+, the vendor has not really bought the top of the page. They have bought entry into a premium queue, where their listing rotates back to the top (every 15 days) with everyone else who also paid to be special.

The value claim starts to wobble from both directions. In the regional suburb, the property may not need the extra prominence. In the metro suburb, the property may pay far more for prominence that is diluted by saturation. Either way, the agent is left explaining a product where the relationship between price, cost and actual marginal value is anything but transparent.

Consumer attention also matters here. Serious buyers will often scroll further than casual browsers, especially when they are actively looking in a specific suburb. But attention is still not evenly distributed down a search page. Nielsen Norman Group’s eye-tracking work on web reading has long shown that users tend to concentrate attention near the top and left of the content area, with earlier lines receiving more attention than later ones. Real estate search is not exactly the same as reading an article or a general search engine, but the basic point still holds: higher placement generally matters because attention drops as people move down the page.

Once a listing falls down inside a crowded premium tier, the benefit of being “premium” becomes much less clean than the brochure version suggests. REA’s own customer material describes Premiere as appearing at the top of search results above Highlight listings, with automatic rotation to the top every 15 days. That means the practical benefit is not a permanent top position. It is a temporary position inside a paid hierarchy.

The strange thing about a saturated depth product is that once enough agents are paying to stand out, the product becomes less about standing out and more about not being pushed further down the page. Perversely, the vendor paying more may be getting less relative value. The higher-priced suburb may have more total buyer traffic, but that attention is divided across more listings, more premium competitors and more paid rotation. The cheaper regional suburb may have less traffic, but also far less competition. So the question becomes unavoidable: is the vendor paying for a genuine advantage, or simply paying not to be buried underneath everyone else who paid?

That is a very different conversation, and it is exactly the kind of conversation a transparent, competitive market would force into the open.

The ACCC context matters, but the wording needs to stay precise. The public record is that in May 2025 REA confirmed it had received a section 155 notice from the ACCC requiring information about certain subscription offerings. Reuters reported the investigation was at an early stage and that the regulator had not yet formed a view. I have not found a public outcome or closure notice since then.

That makes this year’s pricing email feel even more extraordinary. The regulator was already looking at subscription offerings and market power, and less than a year later regional agents (and I assume metro and suburban agents) are receiving notices that include another rise in Premiere+ and a move to charge Standard Land and Rural listings at the Standard listing rate. If the ACCC wanted a live example of why agents and vendors (the 'consumer') are frustrated, this is not exactly subtle.

The industry should stop being polite about this. A company can be successful, profitable, innovative and useful, while still deserving scrutiny for the way it prices a product that has become close to unavoidable. None of that gives it a licence to keep widening the toll road while agents are left explaining the increase and vendors are told to be grateful for the asphalt.

There is an old saying that you can shear a sheep for life, or gut it once. There is another one I like even more: a parasite knows not to kill the host. That is the part I think REA may be misreading, because the short-term pain is obvious but the longer-term gain, if there is one for the industry, is that this level of overreach accelerates the very alternatives that used to look too hard to build. The higher the toll becomes, the more attractive the bypass starts to look.

The bypass is unlikely to arrive as another portal that looks like the old portal. Homely has been around for years. View.com.au was one of the best-capitalised recent attempts to challenge the REA-Domain structure, and it is now set to close in June. At launch, View reportedly had around 900,000 monthly visits, against roughly 53 million for realestate.com.au and 15.5 million for Domain. The gap did not close.

There is a lesson in that. You do not usually beat an incumbent by building a weaker version of the same thing and hoping people change habits out of goodwill. The challenger has to meet the market where it is going, not where the incumbent became powerful.

The Yellow Pages comparison matters because the winning product was not a better directory. The behaviour changed... people stopped thinking “I need the book” and started thinking “I’ll search for it.” The same pattern shows up in other markets as well. Amazon did not beat bookshops by building a slightly larger bookshop, Netflix did not beat Blockbuster by opening better stores, and digital photography did not wait for Kodak to decide when the market should move. In each case, the old model still had brand, habit and revenue for a while, but the discovery layer, or the access layer, moved somewhere else.

For Australian property, the equivalent change may not be one dramatic replacement. It may be buyers finding more stock through Google, asking AI for recommendations in natural language, registering with agency or franchise ecosystems because some listings appear there first, and discovering property-specific pages that are built to be read by search and AI tools. REA can still be large and important inside that market, but the assumption that every serious buyer starts and finishes inside the portal becomes less safe once discovery starts fragmenting.

That is where the agency-owned channel becomes interesting. Cotality’s Decoding 2026 work says 28% of agencies reported listing properties on their own website before publishing them on portals, and Cotality’s own summary language describes it as one in three agencies publishing new listings on their own website first. Ray White AKG is one of the examples, with Avi Khan saying 16% of their sales now happen in the first week, often before a wider launch.

I would be careful not to overstate the causation. That does not automatically mean every one of those sales avoided REA completely. But the signal is still powerful. Agencies are trying to create attention before the expensive portal phase begins. Vendors are asking for value, not just exposure. Serious buyers are asking for early access and the best agencies are starting to see their own website as more than a brochure.

Ray White matters in this discussion because scale changes behaviour. Cotality points to Ray White AKG as one of the businesses using website-first publishing, and Avi Khan says 16% of their sales now happen in the first week, often before a wider launch. Again, that does not prove every one of those sales happened completely away from REA or Domain, but it does point to a more important trend: if buyers learn that meaningful stock appears inside agency or franchise ecosystems before the major portal campaign, the old habit of starting with the portal starts to weaken.

The arithmetic is worth treating carefully, but it is useful. Ray White reported 90,100 sales in FY24 and market share across Australia and New Zealand of 14.2%. If a one-in-six early-sale pattern ever applied across a network of that size (as it does across the Ray White AKG network), that would be about 15,000 sales. At an illustrative average portal listing cost of $1,500, that represents about $22.5 million in potential listing spend sitting in the early-access zone. At $3,000 it is about $45 million, and at $5,000 it is about $75 million. That is not a claim that REA has already lost that revenue, and the public Cotality wording does not prove every early sale happened completely away from the portals. But it shows why even partial bypass behaviour starts to matter when the network is large enough.

Ray White is only one network. The broader Cotality point is that one in three agencies are already publishing on their own websites first. They will not all have Ray White’s scale, database depth or buyer reach, and they may not get the same first-week result. But even a smaller early-sale pattern across a broader slice of the market starts to put real pressure on the economics of the portal model. If a large enough number of agencies can sell even one in 10 of those early listings before the main portal campaign, the money at risk is no longer theoretical, it starts to sit in the tens or hundreds of millions across the market.

This is where the public-company maths becomes more uncomfortable. REA Australia’s revenue was reported at $1.544 billion in FY25, with Australian Residential revenue at $1.156 billion. As an example, a 10% revenue exposure across a high-margin digital listing business is not just a 10% issue if the cost base does not move down with it. The earnings impact can be larger than the revenue impact, and the market does not usually value a slower-growth or declining classifieds business the same way it values a clean growth story. That does not mean REA collapses (and it is not a share call)... but if discovery starts moving and listing revenue starts becoming less automatic, the valuation question changes.

The next phase of competition may not arrive neatly packaged as another portal. It may look more like the search layer itself changing. Some buyers go to Google. Some ask ChatGPT, Claude or Perplexity. Some register for early access with an agency or franchise group. Some find a property-specific page because it answers the question better than a portal filter can. None of those channels replaces REA by itself, but together they start to weaken the old habit of beginning every serious property search in the same place.

Google is a significant pressure point. Similarweb’s April 2026 estimate lists organic search as the top desktop traffic source to realestate.com.au. In other words, a meaningful share of the portal journey still begins outside the portal itself.

Google’s property-listing experiments matter in that context. In selected US markets, Google has tested expanded home listing features directly inside search, including property details, photos, request-tour functions and agent contact pathways. That looks very different to the old model of sending a buyer to a portal link. It is closer to what Google already does with Maps, Shopping and local business results: keep the user inside Google for longer, make the result more useful, and monetise the action more directly.

REA clearly understands some of this, because its ChatGPT app lets users search realestate.com.au listings inside ChatGPT and refine results conversationally. That is useful, but it still feels like a hedge sitting beside the existing model rather than a wholesale rethink of property search. Zillow’s AI Mode appears more ambitious because it is pushing AI-led search and guided discovery deeper into the product itself, including search, comparison and connection to agents. REA may be working on more than we can see publicly, but from the outside, the response still looks lighter than the scale of the threat.

The frustrating part is that REA had plenty of other choices here, if it wanted to meet the market rather than simply push harder into it. Pay on Sale already exists, but the current version is still tightly controlled and still carries a premium. Campaign Flex points can be used for Pay on Sale, but REA says some options still include a 20% premium. That is not a truly market-sensitive rethink. It is a limited flexibility product attached to the same pricing architecture.

A supplier trying to meet the market could have done a lot more with that advantage. A genuine pay-on-success product. A lower upfront rate with a higher fee only if the property sells. A time-based listing fee where a fast sale costs less and a longer campaign costs more. A capped cost-per-enquiry model. A discounted early-launch period designed to compete with agency website-first campaigns. A regional fairness model where low-volume suburbs are priced differently. A proper volume discount. A freeze for one year while the market adjusts to AI, Google, vendor fatigue and regulatory scrutiny.

Instead, agents received another price notice, wrapped in the usual language of flexibility, performance and support.

There is also a market signal sitting underneath this, and I would be careful how far to take it because this is not financial advice and I am not pretending to give a share call. REA’s share price has clearly been under pressure over the past year, although it has also recovered from lows and can still react positively to strong quarterly numbers. That does not prove the thesis. Share prices move for many reasons, and short-term market reactions can be noisy. But it does suggest the market is at least asking whether the old classifieds model is as untouchable as it once looked.

REA can still produce strong operating numbers. The company is still powerful, still profitable, still dominant, and still deeply embedded in the way Australian property is marketed. But a strong quarter does not remove the structural question. If the next phase of competition comes from Google, AI search, agency-owned inventory, website-first listings, franchise ecosystems, Facebook Marketplace, direct buyer databases and listing-specific sites, then REA is no longer only defending against another portal. It is defending the assumption that the portal remains the unavoidable front door to property discovery.

The pricing decision looks risky in that context. If the old model were completely safe, lifting the toll would make more sense. If the road ahead is getting more contested, rebuilding trust with the people who supply the listings and explain the bill to vendors would seem more useful. This pricing notice suggests REA is still acting as though the first version is true.

My view is that pressure is now coming from too many directions to ignore. Agencies are already putting more stock on their own sites before the portals. Ray White’s early-launch strategy is showing that buyers can be trained to look somewhere else first. Google is already a major entry point into the REA journey and has tested richer property listing experiences overseas (which will likely come to Australia). AI search is changing the way people ask for property information. And every price rise gives agents and vendors another reason to try the alternatives more seriously.

REA will not disappear because of one pricing cycle, and that is not really the argument anyway. A platform with that much reach, habit and vendor expectation can remain powerful for a long time. The question is whether the automatic dependence starts to weaken. If more listings appear elsewhere first, if more buyers learn to search before the portal, if Google keeps more of the journey inside search, and if AI starts answering property questions rather than simply pointing people to links, then REA’s pricing power starts to look less permanent than it did a few years ago.

The FY27 notice feels like overreach because REA had a moment here where it could have read the market differently. It could have frozen pricing for a year. It could have offered genuine volume relief. It could have made Pay on Sale a serious product rather than a controlled option with a premium. It could have created a lower-cost early campaign product to compete with website-first launches. It could have given agents something meaningful to take back to vendors other than another increase wrapped in softer language.

Instead, it appears to have widened the toll, which may end up being the mistake. Agents will not suddenly stop using realestate.com.au, but every one of these increases adds energy to the alternatives. It makes Google more interesting. It makes Ray White’s approach more interesting. It makes agency-owned websites more important. It makes AI discovery more valuable. It makes direct buyer databases more strategic. It makes vendors more open to a pre-portal campaign.

There is some light in that, even if the short-term pain is real. Overreach has a way of forcing behaviour that would otherwise take much longer to change. I have been testing this in a small way with listing-specific websites, built for search visibility and AI discovery as part of a broader product direction. It is early and I would not pretend the model is proven yet, but even a small slice of campaign attention matters when it comes from a channel the agency controls. Not because it replaces the portal overnight, but because it changes the conversation.

The same is true of Facebook Marketplace enquiries, direct database activity, early-access campaigns and AI-readable property pages. None of them are the answer on their own. But together they start to show what the next market might look like: fragmented discovery, more direct inventory, earlier buyer engagement, stronger agency-owned channels, and less automatic reliance on a single toll road.

The practical response for agents is not to rant once and then keep doing exactly the same thing. The response is to build more independence while the economics are giving you a reason to do it. Treat your agency website as campaign infrastructure, not a brochure. Treat every listing page as something Google and AI systems need to understand. Treat early access as a real buyer strategy, not just a vague off-market claim. Treat your database as an audience you own, not a storage cupboard. Treat vendor reporting as proof of where the campaign is working. And treat portal spend as a line item that needs to earn its place, not a sacred cost that sits above scrutiny.

REA has built a very strong business. It has genuine reach, real value and a powerful consumer habit, and none of that should be dismissed. It has also built something very close to monopoly power in the Australian residential advertising market, depending on how tightly you define the market, and its behaviour probably says more than any slogan could. A business in a truly competitive market does not keep lifting prices well ahead of value, widen the charge base, hide suburb pricing behind contracts, sell a saturated premium product, and still expect agents to explain the bill to owners year after year. It does that when it believes the customer has nowhere else serious to go.

Audience share is part of that, but the repeated ability to behave this way is the stronger signal. If a supplier can raise prices year after year, charge different amounts suburb by suburb, keep the actual pricing hard to compare, and rely on agents to sell the expense on its behalf, the market is telling you something. The product may have value, but the pricing power is doing a lot of the work.

The problem for REA is that the next phase of competition may not arrive neatly packaged as another portal. It may arrive through buyer behaviour changing at the edges, through more listings appearing on agency websites first, through franchise groups building stronger direct channels, through Google keeping more discovery inside search, through AI making property search less dependent on old portal filters, and through agents building their own digital assets because the economics finally make the effort worthwhile.

That shift is harder to manage than one neat competitor arriving with a familiar logo and a slightly cheaper product. REA can copy a feature, buy a business, adjust a product or undercut a rival if the threat is obvious enough. A gradual weakening of the old assumption that every serious campaign has to begin and end with the same portal pathway is different. REA may still be the largest player for a long time, but if agents and vendors start to believe there is a credible path around part of the spend, the old pricing model begins to look more exposed.

REA’s own pricing may help bring that forward. There is a familiar pattern when the official path keeps getting more expensive. At first, people complain and keep paying. Then some start testing workarounds. Then the workarounds improve because the economics make them worth building. Electricity networks saw a version of that with solar. Governments have seen versions of it with high tobacco excise and black-market supply. Real estate portals are a different category, but the behavioural logic is similar enough: once the price feels out of step with the value, alternatives that once looked marginal start getting a lot more attention.

Seen from that angle, this is not only a story about portal gouging. It is also a story about what gouging eventually produces: a frustrated customer base, a lucrative market for new entrants, and a set of technologies and agency behaviours that are starting to make the bypass feel less theoretical than it did a few years ago.

If REA had used this pricing cycle to show restraint, rethink the product, or give agents something practical to take back to owners, it may have bought itself more goodwill for the change that is coming. Instead, it appears to have pushed harder into the same model. That may work for another year. Maybe longer. But if buyer discovery keeps changing, if agencies keep testing early-access channels, and if vendors keep questioning the portal bill, the old pathway becomes less automatic. That is usually how these things begin: not with one dramatic collapse, but with enough people quietly changing where they look first.